Objectives To screen for cognitive and behavioural impairment in people with amyotrophic lateral sclerosis (ALS) and controls with neuromuscular disease and to correlate these with clinical features.

Methods 108 people with ALS and 60 controls with other neuromuscular diseases were recruited. Addenbrooke’s cognitive examination-III (ACE-III), the Frontal assessment battery (FAB), and the executive function component of the Edinburgh cognitive and behavioural ALS screen (ECAS) were used to assess these people. The Amyotrophic lateral sclerosis-frontotemporal dementia questionnaire (ALS-FTD-Q) and the Motor Neuron Disease Behavioural instrument (MiND-B) were administered to the caregivers of people with ALS. The prevalence of abnormalities was determined and correlated with clinical features and survival. In 37 people with ALS, serial studies were performed.

Results The frequencies of cognitive impairment based on the ACE-III and FAB were 30.0% and 14.0%, in ALS and 11.7% and 3.3% in controls with neuromuscular disease, respectively. Age and years of education influenced the results of the ACE-III and ECAS executive function. In ALS, the frequencies of behavioural impairment based on ALS-FTD-Q and MiND-B were 32.1% and 39.4%, respectively. There was significant correlation of ALS-FTD-Q and MiND-B with the ALSFRS-R score. ALS participants with cognitive impairment measured with ACEIII had significantly shorter survival time than those without (p=0.0276). ALS participants with behavioural impairment measured with ALS-FTD-Q had worse prognosis than those without. No significant difference was found in the first two serial cognitive tests based on ACE-III and FAB by using generalised estimating equation.

Conclusions There was a greater frequency of cognitive impairment in people with ALS than in patients with other neuromuscular diseases. The cognitive and behavioural tests are potential biomarkers of the prognosis of ALS. The results of cognitive tests are stable over 6 months and possibly longer.